Solid Earth, 3, 327–337, 2012
www.solid-earth.net/3/327/2012/
doi:10.5194/se-3-327-2012
© Author(s) 2012. CC Attribution 3.0 License.
Nature of orogenesis and volcanism in the Caucasus region
based on results of regional tomography
I. Koulakov1 , I. Zabelina1 , I. Amanatashvili2 , and V. Meskhia2
1 Institute of Petroleum Geology and Geophysics, SB RAS, Novosibirsk, Russia
2 Ilia State University, Tbilisi, Georgia
Correspondence to: I. Koulakov (koulakoviy@ipgg.nsc.ru)
Received: 16 May 2012 – Published in Solid Earth Discuss.: 7 June 2012
Revised: 10 September 2012 – Accepted: 17 September 2012 – Published: 17 October 2012
Abstract. In the paper, we discuss the problem of con-
tinental collision and related volcanism in the Caucasus
and surrounding areas based on the analysis of the up-
per mantle seismic structure in a recently derived model
by Koulakov (2011). This model, which includes P and S-
velocity anomalies down to 1000 km depth, was obtained
from tomographic inversion of worldwide travel time data
from the catalogue of the International Seismological Center.
It can be seen that the Caucasus region is squeezed between
two continental plates, Arabian to the south and European
to the north, which are displayed in the tomographic model
as high-velocity bodies down to about 200–250 km depth.
On the contrary, a very bright low-velocity anomaly beneath
the collision area implies that the lithosphere in this zone is
very thin, which is also supported by strong horizontal de-
formations and crustal thickening indicating weak properties
of the lithosphere. In the contact between stable continental
and collision zones, we observe a rather complex alternation
of seismic anomalies having the shapes of sinking drops. We
propose that the convergence process causes crustal thicken-
ing and transformation of the lower crust material into the
dense eclogite. When achieving a critical mass, the dense
eclogitic drops trigger detachment of the mantle lithosphere
and its delamination. The observed high-velocity bodies in
the upper mantle may indicate the parts of the descending
mantle lithosphere which were detached from the edges of
the continental lithosphere plates. Very thin, or even absent,
mantle parts of the lithosphere leads to the presence of hot
asthenosphere just below the crust. The crustal shortening
and eclogitisation of the lower crustal layer leads to the dom-
inantly felsic composition of the crust which is favourable
for the upward heat transport from the mantle. This, and also
Published by Copernicus Publications on behalf of the European Geosciences Union.
Solid Earth
the factors of frictional heating and the radioactivity of felsic
rocks, may be the origin of volcanic centres in the Caucasus
and surrounding collisional areas.
1 Introduction
Caucasus is a part of the Alpine-Himalayan orogenic belt
which is the largest continental collision zone in the world. In
the Caucasis segment of the belt, the collision occurs due to
convergence of the Arabian and European continental plates
in a zone located between two basins of presumably oceanic
nature, Black Sea and South Caucasian Plate (Fig. 1). This
collision determines active recent tectonic processes mani-
fested in intensive mountain building, seismicity and Ceno-
zoic volcanism. High level of seismic hazard in this densely
populated region is one of the main reasons for vital interest
to the tectonics of this region.
Mechanism of continental collision is presently not com-
pletely understood and it is actively discussed by specialists
in different domains of geosciences. Considering most exam-
ples of continental collision (e.g., Dewey et al., 1986), one
can see that convergence of continental blocks causes con-
siderable crustal thickening which is roughly proportional
to the value of shortening. At the same time, the fate of
the lithosphere beneath the continental belts remains a dis-
putable topic. While doubling the crust, the collision hardly
results at thickening of the lithosphere: strong thick litho-
sphere would make impossible the observed active deforma-
tions in orogenic belts. Active mountain building and strong
deformations imply that the lithosphere in the collision zones
is weak, and this means that a part of the mantle lithosphere
328 I. Koulakov et al.: Nature of orogenesis and volcanism in the Caucasus region
Fig. 1. Main tectonic units in Caucasus and surrounding areas over-
laid on a shaded relief map. Yellow stars depict the recent volcanoes
in Caucasus (Adamia et al., 2011). Major volcanoes are named with
yellow characters. Areas of Cenozoic volcanism in Iran compiled
from Nezafati (2006), Verdel et al. (2007) are marked with yel-
low. TP is Turan Plate; KD is Kopeth-Dagh; SCB is South Caspian
Basin. White arrow marks the direction of the Arabian Plate dis-
placement.
disappears. However, the details of the mantle lithosphere re-
cycling are still not well understood. Is it subducted to the
mantle similarly as in cases of oceanic subduction or sink
in another way? These questions are actively discussed in
the scientific community (e.g., Dewey and Bird, 1970; Eng-
land and Houseman, 1989; Ershov and Nikishin, 2004). Seis-
mic tomography, which allows imaging the structures at great
depths, is one of the most powerful tools to clarify these and
other geodynamic questions.
Deep seismic structure beneath the Caucasus region has
been investigated in many geophysical studies, mainly us-
ing seismic tomography. Mantle structure beneath Caucasus
and the surrounding areas has been studied using global and
regional seismic modelling based on travel times of body
waves (e.g., Neprochnov et al., 1970; Hearn and Ni, 1994;
Al-Lazki et al., 2004; Gök et al., 2003), surface waves (e.g.,
Maggi and Priestley, 2005; Sandvol et al., 2001; Villasenor et
al., 2001) and seismic attenuation (Sarker and Abers, 1998).
Solid Earth, 3, 327–337, 2012
Most of the studies display generally consistent features
in the lithosphere depth intervals (50–250 km) with higher
seismic velocities in areas corresponding to the continental
blocks and low velocities beneath the folded belt. Here we
base our discussion on a recent seismic model of Asia by
Koulakov (2011) which enables higher frequency features
compared to most of the previously published regional and
global tomographic studies. In order to verify several issues
and to check the reliability of the proposed geodynamic sce-
nario, in this study we provide additional synthetic tests ori-
ented specially to the target region. Based on considering the
tomographic model by Koulakov (2011) we will provide our
answers to the following two questions:
1. What happens to the mantle part of the continental litho-
sphere during the continent-continent collision in Cau-
casus?
2. What is the nature of the active Cenozoic volcanism in
Caucasus and surrounding areas?
2 Geodynamics and volcanism of Caucasus and
surrounding areas
The geological evolution of the Caucasian region (Fig. 1)
mostly controlled by convergence of Eurasian and Africa-
Arabian continental lithosphere plates. According to geode-
tic data, the total rate of the convergence is ∼20–30 mm yr−1
(e.g., DeMets et al., 1990). More detailed analysis of regional
deformations shows that about 60 % of this rate is taken by
the Lesser Caucasian suture, and the rest is accommodated
in crustal shortening in the Southern Caucasus (Allen et al.,
2004; Forte et al., 2010).
Pre-Cenozoic evolution of this area was connected with
the closing of the Tethys Ocean between Eurasian and Gond-
wana continental parts (e.g., Khain 1975; Adamia, 1975;
Adamia et al., 2008; Zakariadze et al., 2007). Convergence
of the ocean with continents during the Late Proterozoic–
Early Cenozoic pre-collisional stage resulted at accretion of
island arcs, intra-arc rifts and back-arc basins etc. Thus, very
large variety of arc volcanism age can be found in the col-
lision zone around Caucasus (Adamia et al., 2011). Fold-
thrust belts in the Great and Lesser Caucasus and, in between,
the Transcaucasian intermontane depression were formed af-
ter definitive closure of Tethys in this segment of the colli-
sion belt during syn-collisional (Oligocene-Middle Miocene)
and post-collisional (Late Miocene-Quaternary) stages of the
Late Alpine tectonic cycle (Adamia et al., 2011). High inten-
sity of the collision processes in Caucasus are possibly due
to its location in a gap between two presumably oceanic type
basins, Black Sea and South Caspian, which are thought to
be the remnant parts of Tethys (see for example, Fig. 21 in
Adamia et al., 2011).
www.solid-earth.net/3/327/2012/
I. Koulakov et al.: Nature of orogenesis and volcanism in the Caucasus region 329

Late Cenozoic intrusive and extrusive volcanism is

widespread throughout the collision zone from Anatolia to
Iran. In Turkey the outcrops of calc-alkali volcanic rocks
are observed along two suture zones in different sides of
the Anatolian peninsula. The southern volcanic branch ex-
tends to Iran along the Urmieh-Dokhtar magmatic arc which
is oriented parallel to the Zagros belt. The northern branch
appears to be adjoining with intensive post-collisional mag-
matic manifestations in Caucasus.
Quaternary volcanism in Caucasus and surrounding ar-
eas is mostly represented by andesites-to-dacite series. The
dacitic lavas were actively erupted in a time period from
760 000 a to 30 000 a in the Javakheti highlands (Lebedev
et al., 2004). Products of the most recent and mostly uplifted
segments of Caucasus including post-collisional volcanoes
of the Elbrus, Chegem and Keli-Kazbegi are represented by
lavas of calc-alkaline subalkaline andesite-basalt, andesite-
dacite rhyolite composition (Tutberidze 2004; Koronovsky
and Demina, 2007). There are many evidences of Pliocene-
Quaternary ages of eruptions for some of these volcanoes, for
example, ∼6000 years near the Kazbegi volcano (Djanelidze
et al., 1982). It is interesting that some authors (e.g., Lebedev
et al., 2008) observe a “dominoes effect” when the magmatic
activity is migrated northward from one volcanic center to
another.
Several geodynamic models have been proposed to ex-
plain the Late Miocene-Quaternary calc-alkaline volcanism
of Caucasus, such as the detachment model of the last piece Fig. 2. Distribution of data from the ISC catalogue: triangles depict
of subducted oceanic lithosphere (e.g., Innocenti et al., 1982) stations, red dots are the events.
or the lithosphere delamination (Pearce et al., 1990; Keskin
et al., 1998). In this paper, we will provide additional argu-
ments for the second concept based on recent tomographic This allowed minimising a problem existing in global stud-
images of the upper mantle. ies when the same damping parameter caused loss of infor-
mation in densely covered areas and artificial instabilities in
parts with insufficient data amount. In the presented model,
3 Tomographic model for each window, the value of damping was separately tuned
to enable the optimal reconstruction of a synthetic model.
This study is based on analysis of P and S velocity models For the selected area, we used four circular windows of 8◦
beneath Asia down to 1000 km depth computed by Koulakov radius with approximately 300 000 rays in total. S-data took
(2011). This model was constructed using arrival times of about 10 % of the total amount. The distribution of stations
seismic body waves reported in the worldwide ISC catalogue and events used in this study is shown in Fig. 2. Number of
(ISC, 2001) in the time period from 1964 to 2007 based the parameterisation nodes in each window varied from 6000
on the tomographic approach developed in Koulakov and to 9000 for the P-data and from 4000 to 5500 for the S-data,
Sobolev, (2006). All the data from the catalogue were ini- depending on the data coverage.
tially reprocessed which resulted at relocation of the events Here we provide several horizontal and vertical sections of
and rejection of large amount of outliers. the considered model shown in Figs. 3, 4 and 5 which corre-
The inversion was performed separately in 32 overlapped spond to the Caucasus region. Note that one should be careful
circular areas which covered most part of Asia. All data with with the interpretation of absolute values of anomalies given
ray paths traveling, at least partly, through the study volume, in this model. The amplitudes of anomalies in seismic tomog-
were considered in this study. This included, the data from raphy studies are strongly affected by damping values which
events located in the study area recorded by the worldwide are used for the inversion. This problem is especially impor-
station network and picks from long-distant events recorded tant in the case of using the ISC data which contain strong
by stations in the study region. It was important that free in- noise. To extract a coherent signal, one should apply strong
version parameters were determined separately for each cir- damping which reduces the amplitudes of retrieved hetero-
cular window based on the results of synthetic modelling. geneities. It can be seen, for example, that in the presented

www.solid-earth.net/3/327/2012/ Solid Earth, 3, 327–337, 2012

330 I. Koulakov et al.: Nature of orogenesis and volcanism in the Caucasus region

Fig. 3. P-velocity anomalies in six horizontal sections, yellow stars and polygons mark the locations of recent volcanoes and folcanic fields
in Caucasus and surrounding areas.

results, the amplitudes of P-anomalies are stronger in shal-
low layers than those of the S-anomalies, and this can be
explained by stronger noise level in S-data and, correspond-
ingly, higher damping used in inversion. In most cases, true
amplitudes cannot be achieved, as the reducing of damp-
ing causes the inversion instability. This is a fundamental
problem which is actual not only for the model considered
here, but for any tomographic studies. This should be taken
into account when interpreting these results on a quantitative
level, and especially, when converting P and S velocities into
petrophysical parameters (temperature, composition, density,
etc.). One of the approaches to estimate the realistic ampli-
tudes of heterogeneities is synthetic modelling simulating re-
alistic patterns, noise level and the main workflow used for
real data processing. Comparison of the initial and recovered
structures gives an idea about reduction of anomalies due to
damping.
The model by Koulakov (2011) has been verified using
many different tests. For example, the contribution of the ran-
dom noise in the data was estimated based on the “odd/even”
test with independent inversions of two data subsets with odd
and even numbers of events. The spatial resolution was eval-
uated using several checkerboard tests with different sizes
of patterns. In order to ground the approach with inversions
in overlapping windows, the synthetic modelling was per-
formed using a model with realistic shapes of structures. The
travel times for this test were computed in the entire area,
whereas the inversion was performed in separate windows.
Here we provide some additional tests. In Fig. 6, we
present a series of checkerboard models: two models

Solid Earth, 3, 327–337, 2012 www.solid-earth.net/3/327/2012/

I. Koulakov et al.: Nature of orogenesis and volcanism in the Caucasus region
Fig. 4. Same as Fig. 3, but for the S-velocity anomalies
for the P-data with the size of 2◦ × 1.5◦ × 300 km
and 1.5◦ × 1◦ × 200 km and two models for the S-data
(3 × 2.5 × 300 km and 2 × 1.5 × 200 km). In all cases, these
models were finer than considered in tests in Koulakov
(2011). When computing the synthetic data, we added the
random noise with the rms of 0.3 s. Furthermore, to simu-
late the existence of blunders in the ISC catalogue, we also
added 5 % of “outliers” for which the noise was multiplied by
ten. As a result, the variance reduction after the inversion for
the synthetic data was about 25–35 % which is significantly
lower than in the case of real data (45–50 %). Despite these
“pessimistic” simulations, it can be seen that in areas with
sufficient amount of data, the checkerboard patterns can be
correctly resolved. For the coarser model, the resolved area
covers the entire central part of the study area. For the finer
board, having a size of about 100 km, the satisfactory recon-
www.solid-earth.net/3/327/2012/
331
struction is achieved only in areas of Turkey, Caucasus and
Zagros; in most other areas the anomalies are not visible. For-
tunately for us, the areas with the highest resolution are the
most attractive from the geological point-of-view and mostly
discussed in the next section.
Another test shown in Fig. 7a consists of the recovering of
a model with realistic shapes of anomalies defined in verti-
cal Sect. 2. The conditions of synthetic modelling were the
same as in the cases of the checkerboard tests. Because of
larger size of synthetic patterns, close to the real anoma-
lies, the value of variance reduction in this case was about
50 %, which is similar to one observed for the real data in-
version. The reconstruction results show that the shapes and
locations of most features are generally correct both for P
and S data. However, at greater depths, where the amount
of rays is much lower, the resolution of recovering is much
Solid Earth, 3, 327–337, 2012
332 I. Koulakov et al.: Nature of orogenesis and volcanism in the Caucasus region
Fig. 5. P- and S-velocity anomalies in three vertical sections. The locations of the profiles are shown in maps in Figs. 3 and 4. Relief along
the profile is shown above each plot.
poorer, the amplitudes of anomalies are much weaker than in
the “true” model and they are strongly smeared, especially
for the S model. For example, for the high-velocity “drops”,
in the reconstruction results we cannot separate them and say
exact the number of anomalies. At the same time, it is impor-
tant for the interpretation that we can detect the existence of
these drops in the mantle, though without resolving their de-
tails. This should be taken into account while constructing a
geodynamic interpretation.
To check the possibility of vertical smearing and leakage
of the crustal anomalies to the mantle, we made another test
shown in Fig. 7b. The configurations of synthetic anoma-
lies are the same as in the previous case, except for the
low-velocity anomaly in the middle part of the profile de-
fined down to 50–60 km which represents the thick crust. It
can be seen that this anomaly is correctly resolved in both
P and S velocities; no vertical leakage is observed and the
lower boundary of the “crust” is reconstructed at the cor-
rect depth. From this test, we can conclude that the low-
Solid Earth, 3, 327–337, 2012
velocity anomaly, which is observed beneath the Caucasus
mountains, really represents the mantle structure.
4 Discussion
In shallower depth sections down to 220 km depth, we can
clearly observe higher P and S velocities associated to the
south with the Arabian plate and to the north with the Euro-
pean plate which consists of several microplates in the con-
tact zone, such as Scythian and Turan plates. Lower seis-
mic velocities are observed beneath the collision zone in the
areas of the major mountain belts. P and S models display
generally consistent structures; however, the amplitude of P
anomalies in the shallower sections is considerably higher.
However, as was discussed in the previous section, this re-
flects rather the damping issues that the real relationships of
amplitudes. It can be seen that all recent volcanic centres ex-
actly fit to the low-velocity patterns of P and S anomalies in
shallower sections.
www.solid-earth.net/3/327/2012/
I. Koulakov et al.: Nature of orogenesis and volcanism in the Caucasus region 333

Fig. 6. Checkerboard tests with different sizes of anomalies for P- and S-models in three horizontal sections. Depths of the sections correspond
to the middle level of the checkerboard patterns. The sizes of synthetic anomalies are indicated above each column.

In vertical sections, we can see that the thickness of high- than estimated by other authors based on different meth-
velocity layers related to the Arabian and European litho- ods (e.g., Artemieva, 2003). However, it should be kept in
spheric plates is about 200–250 km which is a little bit higher mind that the lithosphere related anomalies might be smeared

www.solid-earth.net/3/327/2012/ Solid Earth, 3, 327–337, 2012

334 I. Koulakov et al.: Nature of orogenesis and volcanism in the Caucasus region

Fig. 7. Two synthetic tests with realistic patterns defined in a vertical Sect. 2, same as indicated in Fig. 5. Upper plots show the configurations
of the synthetic models; middle and lower plots are the reconstruction results for the P and S anomalies.

downward due to the limited vertical resolution. In the transi-
tion zones between high-velocities in continental blocks and
low-velocities in the collisional belt, the structure of anoma-
lies is rather complex with alternating high- and low-velocity
anomalies. It can be seen that high-velocity anomalies form
drop-shaped bodies which seem to sink to a greater depth
mantle. Taking into account the results of synthetic test with
vertical anomalies in Fig. 7, we can propose that the real data
inversion smears and reduces the amplitudes of anomalies
at greater depths. Thus, the true amplitudes of anomalies in
these drops might be stronger than observed after inversion
of real data.
The obtained results allow us to propose mechanisms of
the lithosphere recycling due to collision and origin of vol-
canism illustrated in Fig. 8. It can be seen that the Arabian
and Eurasian parts are represented by approximately similar
lithosphere type of about 200–250 km thick. It can be pro-
posed that the lithosphere of these plates have the standard
continental type structure which includes upper felsic (gran-
ite) and lower mafic (basaltic) crustal layers and a rigid man-
tle layer which dominates in total strength of the lithosphere
(e.g., Burov and Diament, 1995). When the plates collide, the
crust thickens in the shortening areas between these plates.
In this case, the lower mafic crustal layer appears at greater
depth. Temperature and press increase lead to phase transfor-
mation of the mafic layer into denser eclogite (e.g., Sobolev
et al., 2006). The drops of eclogite are united into larger
bodies which, after reaching a critical mass, descend to the
mantle. These changes in the lower crust may lead to detach-
ment of the gravitationally unstable mantle lithosphere. The
presence of dense eclogite drops may trigger the lithosphere
detachment and accelerate sinking of separate pieces of the
lithosphere as shown in Fig. 8. As proposed by Burov and
Watt (2006), this may lead to the “crème-brûlé” behaviour
of the mantle lithosphere in the collision zone instead of
“jelly sandwich” rheology which is characteristic for non-
deformed continental lithosphere. This process abruptly de-
creases the total strength of the lithosphere and leads to its

Solid Earth, 3, 327–337, 2012 www.solid-earth.net/3/327/2012/

I. Koulakov et al.: Nature of orogenesis and volcanism in the Caucasus region 335

gest that this factor cannot be ignored in the process of de-

lamination (e.g., Schott et al., 2000) and in the origin of vol-
canism (e.g., Stüwe, 1998). Some contribution in heating can
also be related to radioactivity of felsic rocks.

5 Conclusions

Based on our seismic model, we can conclude that mantle

lithosphere beneath the collision zone of Caucasus and sur-
rounding areas is very thin or absent. That is why this seg-
ment squeezed between rigid Arabian and European blocks
behaves as weak lithosphere and is strongly affected by tec-
Fig. 8. Schematic representation of the delamination mechanism in tonic shortening and orogenesis. Tomography results show
the Caucasus region. The crust is composed of the upper felsic (or-
that the continental lithosphere in the contact with the colli-
ange) and lower mafic (green) parts. Blue areas indicate the mantle
parts of the lithosphere. Background is the distribution of P-velocity
sion area is actively destroyed by the process of delamina-
anomalies in vertical Sect. 2, same as in Fig. 5. Relief along the tion and sink as separate drops. We propose that a big role
profile is shown above the plot. Green ellipses schematically mark in triggering the detachment of the mantle lithosphere and
the possible locations of the eclogite drops which were transformed its more active descending is played by eclogitisation in the
from the lower mafic crust in the shortening zone. Red arrows mark lower crust affected by strong shortening. We believe that
the asthenosphere upwelling. delamination mechanism of such type is the major candidate
for the lithosphere recycling in all continent-to-continent col-
lision zones of the world.
Important conclusion of this research consists in the defi-
fast degradation through active delamination (e.g., Kay and
nition of a collisional type of volcanism which is principally
Kay, 1993). According to this hypothesis, the mantle litho-
different of the intraplate and subduction types of volcanism.
sphere at the edges of the collided continental plates should
We propose that the volcanic activity in Caucasus and sur-
be gradually destroyed and delaminate together with dense
rounding collisional areas is presumably due to direct heat-
eclogite produced in the lower crust. These sinking drops are
ing of the crust from the asthenosphere which is possible due
probably visible as high-velocity bodies in vertical sections
to lack of the mantle lithosphere and thinning of the mafic
of our seismic model beneath the edges of the Arabian and
lower crust layer. This type of volcanism might be expected
European continental parts and beneath the collision zone.
in most areas of continental collision, however, in practice
It is important to mention that a similar processes (tec-
it is observed in a limited number of regions. This might be
tonic shortening – crustal thickening – eclogitisation of the
explained by a hypothesis that the collisional type of volcan-
lower crust that triggers delamination of the mantle litho-
ism occurs only under the condition of complete detachment
sphere) was previously suggested for Central Andes based on
of the mantle lithosphere which is not realised in all cases
petrological arguments (Kay and Kay, 1993) and numerical
of continental collision. However, this topic needs additional
modelling (Sobolev and Babeyko, 2005). Rate of this process
investigations based on thermo-mechanical modelling and
was analysed by Jull and Kelemen (2001).
analysis of geological data in different collisional belts.
After detachment of the mantle lithosphere, it is replaced
by hot asthenosphere which may appear directly beneath the
crust. Note that the crust in the collision zone is mostly com- Acknowledgements. The authors are grateful to topical editor,
posed of thick felsic rocks, whereas the lower mafic crust was Taras Gerya, and two anonymous reviewers for fruitful comments
transformed to eclogite and sank together with the mantle which helped us to improve the content of the paper. The problem
lithosphere. The existence of thick felsic crust is supported of continental collision is studied in the framework of the project
by the observed very low velocities in the collision zone at IP SB RAS #76; questions related to volcanism in collision zones
50 km depth, which can be considered as an integral layer are investigated within ONZ-7.3 Project.
for the crustal properties. It is known that the felsic layer
is composed of mechanically weaker rocks which facilitate Edited by: T. Gerya
circulation of hot materials in the thick crust (e.g., Babeyko
et al., 2002; Babeyko and Sobolev, 2005). This favours ac-
tive heat transport from the asthenosphere to the surface that
explains the existence of active volcanic fields in Caucasus
and surrounding collisional areas. At the same time, addi-
tional heating can come from the frictional effects due to
the strong compressional deformations. Some authors sug-

www.solid-earth.net/3/327/2012/ Solid Earth, 3, 327–337, 2012

336 I. Koulakov et al.: Nature of orogenesis and volcanism in the Caucasus region
References
Adamia, S. A., Chkhotua, T., Kekelia, M., Lordkipanidze, M., Shav-
ishvili, I., and Zakariadze, G.: Tectonics of the Caucasus and
adjoining regions: implications for the evolution of the Tethys
ocean, J. Struct. Geol., 3, 437–447, 1981.
Adamia, S. A., Zakariadze, G., Chkhotua, T., Sadradze, N.,
Tsereteli, N., Chabukiani, A., and Gventsadze, A.: Geology of
the Caucasus: A Review, Turkish Journal of Earth Sciences, 20,
489–544, 2011a.
Adamia, S. A.: Plate tectonics and evolution of the Alpine system:
discussion, Geol. Soc. Am. Bull., 86, 719–720, 1975.
Adamia, S. A., Mumladze, T., Sadradze, N., Tsereteli, E., Tsereteli,
N., and Varazanashvili, O.: Late Cenozoic tectonics and geody-
namics of Georgia (SW Caucasus), Georgian International Jour-
nal of Sciences and Technology 1, 77–107, 2008.
Al-Lazki, A. I., Sandvol, E., Seber, D., Barazangi, M., Turkelli, N.,
and Mohamad, R.: Pn tomographic imaging of mantle lid veloc-
ity and anisotropy at the junction of the Arabian, Eurasian and
African plates, Geophys. J. Int., 158, 1024–1040, 2004.
Allen, M., Jackson, J., and Walker, R.: Late Cenozoic reorganiza-
tion of the Arabia-Eurasia collision and the comparison of short-
term and long-term deformation rates, Tectonics, 23, 1–16, 2004.
Artemieva, I. M.: Lithospheric structure, composition, and thermal
regime of the East European craton: Implications for the sub-
sidence of the Russian Platform, Earth Planet. Sci. Lett., 213,
429–444, 2003.
Babeyko, A. Y. and Sobolev, S. V.: Quantifying different modes of
the late Cenozoic shortening in the central Andes, Geology, 33,
621–624, 2005.
Babeyko, A. Y., Sobolev, S. V., Trumbull, R. B., Oncken, O., and
Lavier, L.: Numerical models of crustal scale convection and
partial melting beneath the Altiplano-Puna plateau, Earth Planet.
Sci. Lett., 199, 373–388, 2002.
Burov, E. B. and Diament, M.: The effective elastic thickness (Te) of
continental lithosphere: What does it really mean?, J. Geophys.
Res., 100, 3905–3928, doi:10.1029/94JB02770, 1995.
Burov, E. B. and A. B. Watts: “The long-term strength of continental
lithosphere: “jelly sandwich” or “crème-brûlé”?”, GSA Today,
16, 4–10, 2006
DeMets, C., Gordon, R., Argus, D., and Stein, S.: Current plate mo-
tions, Geophys. J. Int., 101, 425–478, 1990.
Dewey, J. F. and Bird, J. M.: Mountain belts and new global tecton-
ics. J. Geophys. Res., 74, 2625–2467, 1970.
Dewey, J. F., Hempton, M. R., Kidd, W. S. F., Saroglu, F., and
Şengör, A. M. C.: Short,ening of continental lithosphere: The
neotectonics of eastern Anatolia – A young collision zone, in:
Collision Tectonics, edited by: Coward, M. P. and Ries, A. C.,
London, Geol. Soc. London, 3–36, 1986.
Djanelidze, C., Klopotovskaya, N., and Maisuradze, G.: Holocene
volcanites and lake beds near Goristsikhe settlement, Interna-
tional Union for Quarternary Research (INQUA), 11th Congress
Guidebook for Excursion A-16, 28–31, 1982.
England, P. C. and Houseman, G. A.: Extension during continental
convergence with application to the Tibetan Plateau, J. Geophys.
Res., 94, 17561–17579, 1989.
Ershov, A. V. and Nikishin, A. M.: Recent Geodynamics of the
Caucasus-Arabia-East Africa Region, Geotectonics, 38, 123–
136, 2004.
Solid Earth, 3, 327–337, 2012
Forte, A., Cowgill, E., Bernardin, T., Kreylos, O., and Hamann,
B.: Late Cenozoic deformation of Kura fold-thrust belt, southern
Greater Caucasus, Geol. Soc. Am. Bull., 122, 465–486, 2010.
Gök, R., Sandvol, E., Türkelli, N., Seber, D., and Barazangi,
M.: Sn attenuation in the Anatolian and Iranian plateau
and surrounding regions, Geophys. Res. Lett., 30, 8042,
doi:10.1029/2003GL018020, 2003.
Hearn, T. M. and Ni, J.: Pn velocities beneath continental collision
zones: The Turkish-Iranian plateau, Geophys. J. Int., 117, 273–
283, 1994.
Innocenti, F., Mazzuoli, R., Pasquare, G., Radicati, F., and Villari,
L.: Tertiary and Quaternary volcanism of the Erzurum-Kars area
(Eastern Turkey). Geochronological data and geodynamic evolu-
tion, J. Volcanol. Geoth. Res., 13, 223–240, 1982.
International Seismological Centre, Bulletin Disks 1-9 [CD-ROM],
Internatl. Seis. Cent., Thatcham, United Kingdom, 2001.
Jull, M. and Kelemen, P. B.: On the conditions for lower crustal
convective instability, J. Geophys. Res. 106, 6423–6446, 2001.
Kay, R. W. and Kay, S. M.: Delamination and delamination mag-
matism, Tectonophysics, 219, 177–189, 1993.
Keskn, M., Pearce, J. A., and Mitchell, J. G.: Volcanostratigraphy
and geochemistry of collision-related volcanism on the Erzurum-
Kars Plateau, North Eastern Turkey, J. Volcanol. Geoth. Res., 85,
355–404, 1998.
Khain, V.: Structure and main stages in the tectono-magmatic de-
velopment of the Caucasus: an attempt at geodynamic interpre-
tation, Am. J. Sci., 275-A, 131–156, 1975.
Koronovsky, H. and Demina, L.: Disappeared Pliocene volcanoes
of the Main Caucasian Range and grandiose explosion of the
Kazbegi at the beginning of Pleistocene (North Caucasus) Prob-
lems of Geodynamics, Petrography and Metallogeny of the Cau-
casus, Proceedings of the Scientifi c Sessions, Baki, 92–103,
2007. (in Russian),
Koulakov, I.: High-frequency P and S velocity anomalies in
the upper mantle beneath Asia from inversion of world-
wide traveltime data, J. Geophys. Res., 116, B04301,
doi:10.1029/2010JB007938, 2011.
Koulakov, I. and Sobolev, S. V.: A Tomographic Image of Indian
Lithosphere Break-off beneath the Pamir Hindukush Region,
Geophys. J. Int., 164, 425–440, 2006.
Lavier, L. L. and Steckler, M. S.: The effect of sedimentary cover
on the flexural strength of continental lithosphere, Nature 389,
476–479, doi:10.1038/39004, 1997.
Lebedev, V., Chernishev, I., Dudauri, O., Arakelyants, M., Chugaev,
A., Golzman, Y., and Vashakidze, G.: Geochronology of
Neogene-Quaternary dacitic volcanism of the Javakheti highland
(Lesser Caucasus, South Georgia), Proceedings of Geological In-
stitute of Academy of Sciences of Georgia, 119, 535–544, 2004.
(in Russian)
Lebedev, V., Bubnov, S., Dudauri, O., and Vashakidze, G.:
Geochronology of Pliocene volcanism of Javakheti highland
(Lesser Caucasus). Article 2. Eastern part of the Javakheti high-
land, Regional Geological Correlation, Stratigraphy, Geol. Cor-
rel., 16, 101–123, 2008. (in Russian)
Maggi, A. and Priestley, K.: Surface waveform tomography of the
Turkish-Iranian plateau, Geophys. J. Int. 160: 1068–1080, 2005.
Neprochnov, Y. P., Kosminskaya, I. P., and Malovitsky, Y. P.: Struc-
ture of the crust and upper mantle of the Black and Caspian seas,
Tectonophysics, 10, 517–538, 1970.
www.solid-earth.net/3/327/2012/
I. Koulakov et al.: Nature of orogenesis and volcanism in the Caucasus region
Pearce, J., Bender, J., De Long, S., Kidd, W., Low, P., Guner, Y.,
Saroglu, F., Yilmaz, Y., Moorbath, S., and Mitchell, J.: Genesis
of collision volcanism in Eastern Anatolia, Turkey, J. Volcanol.
Geoth. Res., 44, 189–229, 1990.
Sandvol, E., Al-Damegh, K., Calvert, A., Seber, D., Barazangi, M.,
Mohamad, R., Gok, R., Turkelli, N., and Gurbuz, C.: Tomo-
graphic omaging of Lg and Sn propagation in the Middle East,
Pure Appl. Geophys., 158, 1121–1163, 2001.
Sarker, G. and Abers, G. A.: Deep structures along the bound-
ary of a collisional belt: attenuation tomography of P and S
waves in the Greater Caucasus, Geophys. J. Int., 133, 326–340,
doi:10.1046/j.1365-246X.1998.00506.x, 1998.
Sobolev, S. V. and Babeyko, A. Y.: What drives orogeny in the An-
des?, Geology, 33, 617–620, 2005.
Sobolev, S. V., Babeyko, A. Y., Koulakov, I., and Oncken, O.: Mech-
anism of the Andean orogeny: insight from the numerical mod-
elling, in: The Andes – Active Subduction Orogeny, edited by:
Oncken, O., Chong, G., Franz, G., Giese, P., Götze, H.-J., Ramos,
V. A., Strecker, M. R., and Wigger, P., Frontiers in Earth Sci-
ences, 1, Springer, 513–535, 2006.
www.solid-earth.net/3/327/2012/
337
Stüwe, K.: Heat sources of Cretaceous metamorphism in the Eastern
Alps – a discussion, Tectonophysics, 287, 251–269, 1998.
Tutberidze, B.: Geology and Petrology of Alpine Late Orogenic
Magmatism of the Central Part of Caucasian Segment, Tbilisi
University Publishers, Tbilisi, 1–340, 2004. (in Russian)
Villasenor, A., Ritzwoller, M. H., Levshin, A. L., Barmin, M. P.,
Engdahl, E. R., Spakman, W., and Trampert, J.: Shear velocity
structure of central Eurasia from inversion of surface wave ve-
locities, Phys. Earth Planet. Int., 123, 169–184, 2001.
Zakariadze, G., Dlek, Y., Adamia, S., Oberhansli, R., Karpenko, S.,
Bazylev, B., and Solov’eva, N.: Geochemistry and geochronol-
ogy of the Neoproterozoic Pan-African Transcaucasian Massif
(Republic of Georgia) and implications for island-arc evolution
of the late Precambrian Arabian-Nubian Shield, Gondwana Res.,
11, 97–108, 2007.
Solid Earth, 3, 327–337, 2012